Add uprop_get_sc()

2 files changed, 233 insertions, 0 deletions

diff --git a/gen/data-files b/gen/data-files
index b2c4197..5b7921c 100755
--- a/gen/data-files
+++ b/gen/data-files
@@ -21,6 +21,7 @@ extracted/DerivedLineBreak
 extracted/DerivedNumericType
 extracted/DerivedNumericValues
 PropList
+Scripts
 SpecialCasing
 UnicodeData
 '
diff --git a/gen/prop/sc b/gen/prop/sc
new file mode 100755
index 0000000..40fc39b
--- /dev/null
+++ b/gen/prop/sc
@@ -0,0 +1,232 @@
+#!/bin/sh
+
+#!/bin/sh
+
+set -e
+cd "${0%/*}/../.."
+exec >lib/unicode/prop/uprop_get_sc.c
+
+gawk '
+BEGIN {
+	FS = " *(; *|#.*)"
+
+	map["Adlam"]                  = "ADLM"
+	map["Caucasian_Albanian"]     = "AGHB"
+	map["Ahom"]                   = "AHOM"
+	map["Arabic"]                 = "ARAB"
+	map["Imperial_Aramaic"]       = "ARMI"
+	map["Armenian"]               = "ARMN"
+	map["Avestan"]                = "AVST"
+	map["Balinese"]               = "BALI"
+	map["Bamum"]                  = "BAMU"
+	map["Bassa_Vah"]              = "BASS"
+	map["Batak"]                  = "BATK"
+	map["Bengali"]                = "BENG"
+	map["Bhaiksuki"]              = "BHKS"
+	map["Bopomofo"]               = "BOPO"
+	map["Brahmi"]                 = "BRAH"
+	map["Braille"]                = "BRAI"
+	map["Buginese"]               = "BUGI"
+	map["Buhid"]                  = "BUHD"
+	map["Chakma"]                 = "CAKM"
+	map["Canadian_Aboriginal"]    = "CANS"
+	map["Carian"]                 = "CARI"
+	map["Cham"]                   = "CHAM"
+	map["Cherokee"]               = "CHER"
+	map["Chorasmian"]             = "CHRS"
+	map["Coptic"]                 = "COPT"
+	map["Cypro_Minoan"]           = "CPMN"
+	map["Cypriot"]                = "CPRT"
+	map["Cyrillic"]               = "CYRL"
+	map["Devanagari"]             = "DEVA"
+	map["Dives_Akuru"]            = "DIAK"
+	map["Dogra"]                  = "DOGR"
+	map["Deseret"]                = "DSRT"
+	map["Duployan"]               = "DUPL"
+	map["Egyptian_Hieroglyphs"]   = "EGYP"
+	map["Elbasan"]                = "ELBA"
+	map["Elymaic"]                = "ELYM"
+	map["Ethiopic"]               = "ETHI"
+	map["Georgian"]               = "GEOR"
+	map["Glagolitic"]             = "GLAG"
+	map["Gunjala_Gondi"]          = "GONG"
+	map["Masaram_Gondi"]          = "GONM"
+	map["Gothic"]                 = "GOTH"
+	map["Grantha"]                = "GRAN"
+	map["Greek"]                  = "GREK"
+	map["Gujarati"]               = "GUJR"
+	map["Gurmukhi"]               = "GURU"
+	map["Hangul"]                 = "HANG"
+	map["Han"]                    = "HANI"
+	map["Hanunoo"]                = "HANO"
+	map["Hatran"]                 = "HATR"
+	map["Hebrew"]                 = "HEBR"
+	map["Hiragana"]               = "HIRA"
+	map["Anatolian_Hieroglyphs"]  = "HLUW"
+	map["Pahawh_Hmong"]           = "HMNG"
+	map["Nyiakeng_Puachue_Hmong"] = "HMNP"
+	map["Katakana_Or_Hiragana"]   = "HRKT"
+	map["Old_Hungarian"]          = "HUNG"
+	map["Old_Italic"]             = "ITAL"
+	map["Javanese"]               = "JAVA"
+	map["Kayah_Li"]               = "KALI"
+	map["Katakana"]               = "KANA"
+	map["Kawi"]                   = "KAWI"
+	map["Kharoshthi"]             = "KHAR"
+	map["Khmer"]                  = "KHMR"
+	map["Khojki"]                 = "KHOJ"
+	map["Khitan_Small_Script"]    = "KITS"
+	map["Kannada"]                = "KNDA"
+	map["Kaithi"]                 = "KTHI"
+	map["Tai_Tham"]               = "LANA"
+	map["Lao"]                    = "LAOO"
+	map["Latin"]                  = "LATN"
+	map["Lepcha"]                 = "LEPC"
+	map["Limbu"]                  = "LIMB"
+	map["Linear_A"]               = "LINA"
+	map["Linear_B"]               = "LINB"
+	map["Lisu"]                   = "LISU"
+	map["Lycian"]                 = "LYCI"
+	map["Lydian"]                 = "LYDI"
+	map["Mahajani"]               = "MAHJ"
+	map["Makasar"]                = "MAKA"
+	map["Mandaic"]                = "MAND"
+	map["Manichaean"]             = "MANI"
+	map["Marchen"]                = "MARC"
+	map["Medefaidrin"]            = "MEDF"
+	map["Mende_Kikakui"]          = "MEND"
+	map["Meroitic_Cursive"]       = "MERC"
+	map["Meroitic_Hieroglyphs"]   = "MERO"
+	map["Malayalam"]              = "MLYM"
+	map["Modi"]                   = "MODI"
+	map["Mongolian"]              = "MONG"
+	map["Mro"]                    = "MROO"
+	map["Meetei_Mayek"]           = "MTEI"
+	map["Multani"]                = "MULT"
+	map["Myanmar"]                = "MYMR"
+	map["Nag_Mundari"]            = "NAGM"
+	map["Nandinagari"]            = "NAND"
+	map["Old_North_Arabian"]      = "NARB"
+	map["Nabataean"]              = "NBAT"
+	map["Newa"]                   = "NEWA"
+	map["Nko"]                    = "NKOO"
+	map["Nushu"]                  = "NSHU"
+	map["Ogham"]                  = "OGAM"
+	map["Ol_Chiki"]               = "OLCK"
+	map["Old_Turkic"]             = "ORKH"
+	map["Oriya"]                  = "ORYA"
+	map["Osage"]                  = "OSGE"
+	map["Osmanya"]                = "OSMA"
+	map["Old_Uyghur"]             = "OUGR"
+	map["Palmyrene"]              = "PALM"
+	map["Pau_Cin_Hau"]            = "PAUC"
+	map["Old_Permic"]             = "PERM"
+	map["Phags_Pa"]               = "PHAG"
+	map["Inscriptional_Pahlavi"]  = "PHLI"
+	map["Psalter_Pahlavi"]        = "PHLP"
+	map["Phoenician"]             = "PHNX"
+	map["Miao"]                   = "PLRD"
+	map["Inscriptional_Parthian"] = "PRTI"
+	map["Rejang"]                 = "RJNG"
+	map["Hanifi_Rohingya"]        = "ROHG"
+	map["Runic"]                  = "RUNR"
+	map["Samaritan"]              = "SAMR"
+	map["Old_South_Arabian"]      = "SARB"
+	map["Saurashtra"]             = "SAUR"
+	map["SignWriting"]            = "SGNW"
+	map["Shavian"]                = "SHAW"
+	map["Sharada"]                = "SHRD"
+	map["Siddham"]                = "SIDD"
+	map["Khudawadi"]              = "SIND"
+	map["Sinhala"]                = "SINH"
+	map["Sogdian"]                = "SOGD"
+	map["Old_Sogdian"]            = "SOGO"
+	map["Sora_Sompeng"]           = "SORA"
+	map["Soyombo"]                = "SOYO"
+	map["Sundanese"]              = "SUND"
+	map["Syloti_Nagri"]           = "SYLO"
+	map["Syriac"]                 = "SYRC"
+	map["Tagbanwa"]               = "TAGB"
+	map["Takri"]                  = "TAKR"
+	map["Tai_Le"]                 = "TALE"
+	map["New_Tai_Lue"]            = "TALU"
+	map["Tamil"]                  = "TAML"
+	map["Tangut"]                 = "TANG"
+	map["Tai_Viet"]               = "TAVT"
+	map["Telugu"]                 = "TELU"
+	map["Tifinagh"]               = "TFNG"
+	map["Tagalog"]                = "TGLG"
+	map["Thaana"]                 = "THAA"
+	map["Thai"]                   = "THAI"
+	map["Tibetan"]                = "TIBT"
+	map["Tirhuta"]                = "TIRH"
+	map["Tangsa"]                 = "TNSA"
+	map["Toto"]                   = "TOTO"
+	map["Ugaritic"]               = "UGAR"
+	map["Vai"]                    = "VAII"
+	map["Vithkuqi"]               = "VITH"
+	map["Warang_Citi"]            = "WARA"
+	map["Wancho"]                 = "WCHO"
+	map["Old_Persian"]            = "XPEO"
+	map["Cuneiform"]              = "XSUX"
+	map["Yezidi"]                 = "YEZI"
+	map["Yi"]                     = "YIII"
+	map["Zanabazar_Square"]       = "ZANB"
+	map["Inherited"]              = "ZINH"
+	map["Common"]                 = "ZYYY"
+
+	print "/* This file is autogenerated by gen/prop/sc; DO NOT EDIT. */"
+	print ""
+	print "#include \"__bsearch.h\""
+	print "#include \"macros.h\""
+	print "#include \"rune.h\""
+	print "#include \"unicode/prop.h\""
+	print ""
+}
+
+/^[^#]/ {
+	n = split($1, a, /\.\./)
+	lo = strtonum("0X" a[1])
+	hi = strtonum("0X" a[n])
+
+	for (i = lo; i <= hi; i++) {
+		gsub(/^; /, "", $2)
+		props[i] = "SC_" map[$2]
+	}
+}
+
+END {
+	print "static constexpr enum uprop_sc lookup_lat1[] = {"
+	for (i = 0; i < 0x100; i++) {
+		if (i % 8 == 0)
+			printf "\t"
+		printf "%-7s,%s", props[i] ? props[i] : 0, i % 8 == 7 ? "\n" : " "
+	}
+	print "};"
+	print ""
+
+	print "static const struct {"
+	print "\trune lo, hi;"
+	print "\tenum uprop_sc val;"
+	print "} lookup[] = {"
+
+	for (i = 0x100; i <= 0x10FFFF; i++) {
+		if (!props[i])
+			continue
+		lo = i
+		while (props[lo] == props[i + 1])
+			i++
+		printf "\t{RUNE_C(0x%06X), RUNE_C(0x%06X), %s},\n", lo, i, props[i]
+	}
+
+	print "};"
+	print ""
+	print "__MLIB_DEFINE_BSEARCH(enum uprop_sc, lookup, SC_ZZZZ)"
+	print ""
+	print "enum uprop_sc"
+	print "uprop_get_sc(rune ch)"
+	print "{"
+	print "\treturn ch <= lengthof(lookup_lat1) ? lookup_lat1[ch] : mlib_lookup(ch);"
+	print "}"
+}
+' data/Scripts | sed 's/\s*$//'